Activity Number:
|
291
- SPEED: Biopharmaceutical Statistics
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 1, 2017 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #323061
|
View Presentation
|
Title:
|
Power Comparison of Tests of Restricted Mean Survival Time with Log-Rank Test and Generalized Wilcoxon Test Under Various Survival Distributions
|
Author(s):
|
Musashi Fukuda* and Yutaka Matsuyama
|
Companies:
|
Astellas Pharma Inc. and Department of Biostatistics, School of Public Health, The University of Tokyo
|
Keywords:
|
restricted mean survival time ;
log-rank test ;
generalized Wilcoxon test ;
non-proportional hazard ;
power
|
Abstract:
|
Restricted mean survival time (RMST) is attracting attention as an easily interpretable effect measures both with and without proportional-hazard (PH) assumption. However operational characteristics of RMST as a test statistic compared with log-rank test or other rank-type tests are unclear. In this research, we evaluate the power of Wald test of RMST difference and ratio in comparison with log-rank test and generalized Wilcoxon test under various survival distributions with and without PH assumption. We simulate PH-effect, early-effect and delayed-effect type of clinical trials under the Weibull models, piecewise-exponential models, log-logistic-accelerated failure time models and location-scale models. We found the power of RMST was similar to that of generalized Wilcoxon test under most of all scenarios.
|
Authors who are presenting talks have a * after their name.